Adolescents are at high risk for a number of negative health outcomes associated with early and unsafe sexual activity, including infection with human immunodeficiency virus (HIV), other sexually transmitted diseases (e.g., herpes, syphilis), and unintended pregnancy (Center for Disease Control (CDC; 2005). Youth Risk Behavior Survey data suggest more than one-third (34%) of high school students reported having had sexual intercourse, with 6% having initiated before the age of 13. Early initiation of sexual intercourse is more prevalent among males, with 9% of males compared to 4% of females reporting having had sexual intercourse before the age of 13. Beyond early initiation of sexual intercourse, 37% report not using a condom during last sexual intercourse, putting adolescents at significant risk of pregnancy and sexually transmitted diseases. The proposed dissertation integrates individual, familial, biological, and social factors believed to contribute to adolescent sexual risk behaviors into a conceptual model that simultaneously considers the influence and complexity of their combined effects on couple sexual risk behaviors. Specifically, it examines how parental relationship satisfaction, pubertal timing, perceived pubertal timing relative to one's peers, and depression impact sexual risk taking in adolescent couples. Assessments of these variables are made on each couple member separately and then these variables are used to predict the sexual activity of the couple. Four plausible models representing couple dynamics with respect to depression can be specified. The first model, termed the female influence model, states that depression will impact the sexual activity of a couple, but that the primary influence on such behavior is the depression level of the female member of the dyad as opposed to the male member of the dyad. The second model, termed the male influence model, states that male depression levels rather than female depression levels will be most predictive of couple sexual activity. The third model, termed the shared influence model, conceptualizes the depression of both the male and female partner as being independent predictors of couple sexual risk behaviors. The final model, termed the configural influence model, involves more complex couple dynamics. This model is a variant of the shared influence model in that it accounts for risky sexual couple behaviors using the depression levels of both partners but allows for the potential of configural influence. In addition to the independent effects of the depression of both couple member, the configural influence model posits an interaction between each respective member's depression levels such that as female depression increases, male depression becomes more strongly related to risky sexual behaviors. The present dissertation will test these four models and thereby help to elucidate the role of depression on adolescent sexual activity relative to past research. The multi-analytic strategies used in the analysis of the data are described in detail under the Data Analysis Plan. [unreadable] [unreadable] [unreadable] [unreadable] [unreadable] [unreadable]